INTRODUCTION
Research motivation
The E-Learning model emerged in the US in 1999, revolutionizing education by enabling online interaction through electronic communication Significant growth occurred in 2010 with the rise of mobile applications and social networks, enhancing accessibility and interoperability for users worldwide By 2017, the number of E-Learning participants surged from 36 million in 2015 to nearly 70 million, contributing to the industry's remarkable revenue growth According to a 2018 report from the ITAM conference in Vietnam, global E-Learning revenue soared from $51.5 billion in 2016 to over $100 billion in 2017, solidifying its status as one of the world's most dynamic industries.
E-Learning is the most active in the US, the country with the world's top education
In 2018, statistics from the Cyber Educational Institution revealed that over 80% of universities in the country adopted E-Learning as a training method Prominent businesses like Coursera, edX, and Udacity offer Massive Open Online Courses (MOOCs), making online education more accessible Additionally, E-Learning is recognized as an effective tool for employee training, with 77% of companies in the US integrating E-Learning courses into their staff development and retraining initiatives.
Besides the US, Asia is also a market that offers quite "E-Learning" services According to the University World News, the region's total E-Learning revenue in
In 2018, the global E-Learning market was valued at approximately $12.1 billion, with India and China dominating, collectively accounting for 70% of the capital and 30% of all online educational users worldwide China's E-Learning sector generated $5.2 billion in revenue, despite experiencing economic challenges due to the US-China trade war Nevertheless, the E-Learning industry in China continues to thrive, compensating for declines in domestic manufacturing India, while not reaching China's revenue levels, reported $0.7 billion in E-Learning revenue, with BYJU emerging as a key player This startup, which offers online learning applications for K-12 students, has gained significant traction since its launch in October 2015, covering over 1,700 cities and achieving more than 15 million app downloads By March 2018, BYJU's revenue reached $85 million, attracting over $245 million in investments since 2016.
Vietnam is considered to be quickly catching up with the world trend because in
Since 2010, E-Learning has emerged as a global trend, prompting enterprises in Vietnam to explore and launch various online learning platforms such as Violet.vn, Hocmai.vn, and Topica Today, online education has become a popular model, particularly in bustling cities like Ho Chi Minh City, Hanoi, and Danang, attracting a diverse range of users from all educational levels The primary focus of online education in Vietnam includes international language classes, catering to the growing demand for accessible learning options.
Test preparation courses and specialized knowledge lectures (level 2 and level 3) offer a diverse range of skill-building opportunities E-Learning lectures feature rich content presented through various engaging formats, including videos, audio clips, and vivid illustrations, while maintaining interaction with instructors Notably, many online learning platforms provide high school students with access to extensive databases of lectures that align closely with the curriculum set by the Ministry of Education.
In the first quarter of 2019, Vietnam's labor market revealed over one million unemployed individuals, including 124,500 with university degrees, highlighting a significant demand for foreign language and soft skills training to enhance job competitiveness Many students and employees prefer online courses due to their convenience, allowing them to balance work and study effectively Additionally, the affordability of online learning makes it a popular choice among students, as it is generally less expensive than traditional classroom education.
Hocmai.vn, a well-known online education platform in Vietnam, has successfully attracted 3.5 million members and experiences over 10,000 concurrent visits Offering more than 1,000 courses and 30,000 lectures annually, the platform features content from over 200 educators accessible on various devices, including computers, laptops, and smartphones In the past three years, the number of new users registering for hocmai.vn's online learning system has grown by nearly 20% annually, with a remarkable 30% increase in the percentage of paying learners This indicates a strong upward trend in the adoption of E-Learning among users of hocmai.vn.
Topica Educational Technology Complex, alongside hocmai.vn, is a popular choice for thousands of users seeking diverse online education solutions, including the Online Bachelor Program (Topica Uni), English language learning (Topica Native), and a technology platform for various online courses (Edumall) Recently, Alipay partnered with Napas to connect their networks, facilitating the rapid influx of hundreds of Asian companies, particularly from Japan, into the Vietnamese marketplace, supported by investments from both domestic and foreign financial firms.
E-Learning is gaining significant traction in Vietnam's education sector, attracting both business investments and governmental support The Ministry of Education and Training has partnered with various companies to implement E-Learning initiatives, including the "Designing E-Learning Lectures" contest in the 2009-2010 school year and online competitions like the Violympic math contest and the Go-English Olympic Games Many universities, such as Nguyen Tat Thanh University and Polytechnic University, are integrating E-Learning into their curricula alongside traditional teaching methods Additionally, FUNiX, part of the FPT Education system, is emerging as a notable player in the online training landscape in Vietnam.
Vietnam is emerging as a promising market for E-Learning, driven by over 60% of the population using the Internet, predominantly young users with significant learning needs Education spending constitutes 5.8% of GDP and 20% of the national budget, according to the Ministry of Education and Training This has led to increased participation in the E-Learning sector, not only from established players but also from numerous Vietnamese startups and foreign investments, particularly from Singapore By the end of 2016, Vietnam had recorded 309 funding ventures in E-Learning, amassing over $767 million in reported equity, with expectations for continued capital inflows in the coming years.
In November 2018, Northstar Singapore Group made a significant investment of up to $50 million in Topica Edtech Group, marking the largest funding for an educational technology company in Southeast Asia Following this, Kaizen Private Equity, an educational hedge fund based in Singapore and India, announced a $10 million investment in Yola, a Vietnamese startup focused on e-learning English education, in August 2019 Shortly after, Everest Education, another Vietnamese educational startup, successfully raised $4 million from Hong Kong-based Hendale Capital to enhance its academic training centers in Ho Chi Minh City, utilizing a blended learning approach that combines traditional classroom instruction with online learning methods.
The E-learning market in Vietnam has experienced significant growth, positioning the country among the top 10 rapidly developing Asian nations in this sector, thanks to the involvement of both domestic companies and foreign investors, as reported by University World News in 2017.
2017, Vietnam was ranked as the nation with the fastest growth rate (about 44.3%) of e- learning, 4.9% larger than Malaysia - an already inherently accelerated country which is strong growth in this area
The COVID-19 pandemic, declared by the WHO, has significantly impacted online learning in Vietnam since the first cases were reported on January 30, 2020 In response to the crisis, the Vietnamese government closed all educational institutions, prompting a shift to online teaching across all levels, from primary to tertiary This transition also led many language and soft skills training centers to offer online courses Globally, institutions like Harvard University responded by launching over 60 free online courses, enhancing accessibility to education during these challenging times.
During the pandemic, residents were urged to stay indoors, leading to a significant increase in the demand for online learning across various fields, including science, programming, social sciences, anthropology, health, and the arts With more free online courses available from institutions like the British Council, Microsoft, and Yale University, individuals found ample opportunities to enhance their knowledge during their time at home This shift towards online education is expected to drive substantial advancements in both the educational and economic sectors, continuing even beyond the pandemic.
Despite Vietnam's potential for online learning growth, a significant number of students drop out of online courses due to a lack of quality assurance This raises the critical question of what factors influence students' decisions to either continue or discontinue their online studies Understanding these influences is essential for improving course retention rates in Vietnam's online education landscape.
Research Objectives
As mentioned in the practical and theoretical motivation above, there are 3 research objectives:
- Investigate the factors affecting learners ‟s intention to persist in e-learning courses in Vietnam
- Investigate how those factors affect learners ‟s intention to persist in online learning courses in Vietnam
- Propose suggestions and solutions for high rate of dropout to attract more learners and keep them to persist with online courses also improve online learning business performances.
Research Questions
Base on research objectives listed above, there are two research questions:
Demographic characteristics such as age, educational level, and gender, along with external factors like family and workplace support, play a crucial role in influencing learners' intentions to persist in e-learning courses in Vietnam Additionally, internal factors, including academic locus of control and overall satisfaction, significantly impact this relationship Understanding how these elements interact can provide valuable insights into enhancing e-learning persistence among Vietnamese learners.
- What aspects and recommendation for e-learning business in order to keep and get more learners, improve online learning business performances?
Research methodology
This thesis used primary data sources collected by the surveys to figure out the kernels aspects can influence the intention to persevere in online courses in Vietnam
Quantitative analysis was applied in this research In that way, primary data collection is possible proceed as follows:
A survey conducted by Google Forms targeted Vietnamese individuals aged 20 and older who were enrolled in online courses, such as language, soft skills, Coursera, and professional training, during the survey period from April 5 to April 20, 2020 The survey was exclusively distributed through online channels.
Information gathered from the study will be circulated investigation and taking care of in the accompanying request:
- Input: The questionnaire was entered using data entry software for research to avoid bias
To ensure data accuracy, it is essential to clean the data through a two-step process First, errors are identified by checking if they fall within the acceptable range Next, any discrepancies found in the data file are corrected, ensuring the integrity of the information.
Information gathered has examined with Amos 24 and SPSS 22 which combine:
- Descriptive statistics analysis: using descriptive statistics analysis to unmitigated analyze in age, educational level and gender
- Reliability analysis: use Cronbach‟s alpha and exponential factor analysis to test the reliability
Research structure
This study develops a model to identify the factors influencing learners' intentions to persist in e-learning courses in Vietnam It is structured into five key chapters: an introduction, a literature review, research methodology, findings, and a conclusion.
This chapter gives an introduction of research motivation, research object, research question, research methodology, research structure.
LITERATURE REVIEW
Definition
Distance education is defined by many scholars as a teaching method where the majority, if not all, of the educational process occurs with a separation between teachers and learners in both time and space.
Distance education is defined as "institution-based, formal education where the learning group is separated, and interactive telecommunications systems connect learners, resources, and instructors" (Schlosser, 2009) This widely accepted definition highlights the essential elements of distance learning, which can be effectively summarized by four key components as outlined in the Encyclopaedia Britannica 2009 Book of the Year.
Distance learning is facilitated through structured educational systems rather than being a solitary study experience While institutions may offer traditional classroom instruction, they must adhere to specific qualifications to be acknowledged as legitimate educators.
Geographical segregation remains a challenge in distance learning, compounded by the impact of varying time zones on both students and teachers However, this mode of education offers significant advantages in terms of accessibility and convenience By implementing well-structured programs, it is possible to bridge the cultural, intellectual, and social gaps among students, fostering a more connected learning environment.
Telecommunication serves as a crucial link between educators and learners, facilitating interaction through various communication methods While electronic forms such as email remain prevalent, traditional communication methods like postal services also contribute to this connection Regardless of the approach—modern or traditional—effective communication is essential for meeting the needs of distance education.
The advancement of communication technologies, including the internet, mobile phones, and email, has significantly reduced the impact of physical distance in education These tools enable students to access materials and documents quickly, fostering a more efficient distance learning experience.
Distance learning, similar to traditional classrooms, creates a study group or learning community that includes teachers, students, and various educational resources such as slides, audio, video, and graphics This collaborative environment enables students to access instructional materials effectively.
Distance learning primarily caters to adult students pursuing university programs, rather than elementary or high school learners These adults often opt for online education to obtain degrees that enhance their career prospects and facilitate promotions This flexible learning approach allows them to engage in lifelong learning without the constraints of full-time study commitments or geographical limitations (Columbaro, 2009).
Distance education offers numerous advantages, primarily by enhancing accessibility to the national education system for individuals facing time, location, financial, or age constraints This flexible learning format allows students to tailor their study schedules to their personal circumstances, saving time and eliminating the need for commuting Additionally, as knowledge evolves rapidly, professionals can stay current without traveling to specialized training programs Online courses also enable instructors to teach from home, contributing to overall cost savings by reducing travel expenses for students.
13 well as lecturers can be saved Different institutions may have the same lecturer, with online courses lecturers can teach more people
While distance learning offers numerous benefits, it also presents significant limitations, with the quality of education being a primary concern This quality is often influenced by the attitudes of management and instructors; when distance learning is perceived as a lesser form of education, it can lead to a lack of effort in adapting training programs Additionally, the absence of direct interaction between lecturers and students makes it challenging to adjust teaching methods in real-time, potentially negatively impacting learning outcomes and student engagement.
Distance learning presents unique challenges for students, as not all individuals are suited for this mode of education Those pursuing online studies may struggle with technology, science, or quantitative subjects Successful distance learners typically possess key personality traits, such as a willingness to accept ambiguity and flexibility Compared to traditional students, they must excel in focus, time management, and technology use, while also being capable of both independent and collaborative work However, the independent nature of distance learning can lead to decreased focus and a fear of learning Communication primarily through the internet and phones may result in incomplete understanding of lessons and feelings of boredom This isolation can diminish learners' activity levels, confidence, and motivation.
The term "dropout" lacks a precise definition, but some authors emphasize positive student behaviors in their interpretations According to Levy (2007), individuals may choose to drop out of an e-learning course due to incurring additional penalties after a period of inactivity or extra fees Additionally, Liu (2009) defines dropout simply as the failure to complete a course, which can result in an incomplete grade or a final "F" on the student's record.
Persistence can be understood as the willingness of learners to keep on going and continue to accomplish studying ambition, or to continue/remain to involve in academic courses (Shin, 2003)
Persistence is often viewed as the antithesis of attrition, representing a continuum of elements that lead to the successful completion of a program (Park, 2009) The term "persistence" gained prominence in 1980, particularly in the context of secondary education, where it was defined as the act of departing from traditional universities (Greer).
In my study, persistence is defined as a multifaceted phenomenon crucial for the completion of online learning Drawing on Berger's (1998) concept of returning intention, I measure students' perseverance by their intention to complete the course they are enrolled in at the time of the survey.
Research Model Literature Review
Over the past few decades, numerous theoretical models of student perseverance have been developed and studied globally One of the earliest frameworks is the psychological model, which includes the Theory of Reasoned Action proposed by Ajzen in 1975 This theory suggests that students' decisions to persist in their studies are largely influenced by their attitudes and beliefs regarding their educational goals.
15 prior actions, perceptions and expectations which become real behavior by the development of an intention to study
Figure 2-1 Theory of reasoned action (Ajzen, 1975)
Volition serves as a crucial intermediary between learning intentions and behaviors, as identified by Corno (1993) It encompasses the behaviors and emotions that help maintain focus on achieving specific goals despite distractions According to Heckhausen (1985), volition is a psychological state that plays a significant role in converting aspirations into actionable steps, highlighting the importance of self-regulation in fostering perseverance While motivation is a key factor that encourages students to engage in their studies, those enrolled in demanding courses may experience decreased motivation when faced with challenges Therefore, understanding the volition process is essential for analyzing student perseverance.
Subsequent psychological models were developed to analyze student persistence and dropout rates, focusing on observations of students and the impact of external factors and campus life on social inclusion Notably, the models proposed by Tinto (1975, 1987, 1993) and Bean (1985) have been widely utilized to elucidate these phenomena.
Tinto's model (1975, 1987, 1993) illustrates the factors influencing student perseverance within the student-institution relationship He posits that academic success hinges on two key components: (1) individual personality traits, particularly those of university students, and (2) experiences gained during university While pre-university experiences and student personality are considered input variables largely unaffected by the institution, the experiences acquired post-enrollment—termed "integration" variables—are significantly influenced by school policies and environment.
Research by Tinto (1987) indicates that stronger involvement in the core activities of institutional life enhances a student's likelihood of persistence Additionally, Pascarella (1991) found that academic convergence during lower secondary school significantly influences a student's perseverance Consequently, when evaluating a student's fit for a school, perseverance often serves as a key criterion.
Figure 2-2 Conceptualization of (Tinto, 1975,1987,1993) ‘s student integration model
Tinto's student convergence model emphasizes that effective student integration relies on shared interactions and the intellectual knowledge acquired in school He uses GPA as a measure of academic integration and considers factors like participation in extracurricular activities and relationships with teachers and peers to evaluate social integration Tinto's research indicates that the synergy between academic and social integration fosters stronger connections among students and increases their persistence in school Additionally, an individual's commitment to their educational goals, alongside their dedication to the institution, plays a crucial role in determining their likelihood of completing their studies.
18 quit school or not and the type of dropout behavior that individuals express Thus, we can predict the lower the commitment to the goal, the stronger the dropout rate
Tinto emphasized that students are at risk of dropping out if they feel disconnected from their peers and teachers, particularly when their values differ from those around them This lack of cultural alignment can lead to feelings of loneliness and a diminished sense of belonging He noted that students may perceive alternative uses of their time and resources as more beneficial than remaining in college, leading to potential dropout Conversely, when students successfully integrate social experiences with their academic pursuits, they are more likely to complete their degrees Therefore, fostering a supportive learning community is crucial for encouraging student retention and success.
Schools play a crucial role in facilitating the integration of students by providing orientation programs that introduce them to the school's culture, regulations, facilities, and community connections These initiatives help both new and returning students adapt more quickly, fostering a sense of belonging and increasing their perseverance Research indicates that students who engage in summer transition programs experience positive outcomes in their integration process (Wolf-Wendel, 1999).
Tinto's model is primarily designed for analyzing the persistence of traditional university students, making it less applicable to non-traditional learners, including undergraduate and postgraduate students as well as adult learners The model overlooks significant external factors that influence students' perceptions and motivation, which are critical for fostering persistence in diverse educational contexts.
2.2.3 Bean and Metzner’s student attrition model
Non-traditional students, as defined by Bean (1985), are typically individuals over the age of 24 who do not reside in dormitories and may attend school part-time Unlike traditional students, they are less influenced by their institution's environment and are primarily focused on the qualifications they achieve through their courses Online courses cater to both traditional and non-traditional students, with a noticeable emphasis on the latter This shift necessitates a reevaluation of persistence criteria in learning, particularly for non-traditional students, who often face unique challenges related to social integration Due to commitments such as family responsibilities and the need to balance work and study, non-traditional students frequently find it difficult to engage in extracurricular activities, as noted by Graham (2000).
Bean (1985) developed a model, drawing on Tinto's work (1975, 1987, 1993) and various psychological theories, to elucidate the persistence of non-traditional students This model highlights that older students rely more on external support systems, such as family, friends, and work, compared to younger students, who typically engage more with school-based support from teachers and peer groups Consequently, the frameworks established by Bean and Metnez offer a more comprehensive understanding of student perseverance in online learning environments.
Figure 2-3 Conceptualization of (Bean, 1985) student attrition model
Bean and Metzner's models, like Tinto's, emphasize the role of the student institution in predicting student perseverance, but they are particularly applicable to non-traditional students The model identifies four key elements influencing persistence: (1) academic variables, including study habits, absenteeism, program fit, and course availability; (2) student background variables, such as age, educational goals, residence status, ethnicity, and prior GPA; (3) environmental factors that impact student experiences; and (4) institutional variables that shape the educational environment.
Numerous factors influence students' academic persistence, including working hours, family responsibilities, financial constraints, external support, and transfer opportunities Additionally, academic outcomes such as GPA, satisfaction, and commitment to goals play a significant role Notably, students facing financial hardships often struggle to maintain their academic focus These variables, largely beyond the school's control, can create significant pressure on students as they attempt to balance work and study commitments.
A study conducted at a suburban university in southeast Virginia (Parker, 1997) highlighted that financial issues are the primary factor affecting non-traditional students' persistence, followed by family responsibilities, work schedule conflicts, and poor academic performance This finding supports the Bean and Metzner model, which posits that favorable academic and environmental conditions promote student persistence, while unfavorable conditions lead to higher dropout rates (Henry, 1993) Specifically, when academic performance is strong but environmental factors are negative, the positive impact of academic success diminishes, potentially leading to dropout, even among high-achieving students, due to low satisfaction, commitment, and various life stressors.
A study by Ashar (1993) highlights that smaller classes emphasizing expertise lead to better social integration and lower absenteeism among non-traditional students compared to larger, less integrated classes This finding supports Tinto's model, which identifies social integration as a key factor influencing the persistence of non-traditional students Consequently, integrating Tinto's and Bean's models offers a more precise understanding of the factors that contribute to the perseverance of non-traditional students.
Research Hypothesis
On the basis of the literature reviews above, the discussions for the hypothesis in this paper can be analyzed as follows:
Hypothesis 1: Demographic characteristics has an indirect effect on intention to persist through internal factors
A study by Rovai (2001) revealed gender differences in communication styles and community engagement in online courses The research indicated that men tended to express themselves with a more autonomous voice, while a significant majority of women demonstrated a different approach to communication and community awareness.
Research indicates that individuals with a strong sense of community often employ a connected voice in their written communications, while those with a weaker sense of community tend to use a more independent voice This distinction is particularly relevant in the context of e-learning, as a low sense of community can lead to feelings of isolation and difficulty in forming connections, which negatively impacts student perseverance in online courses Additionally, studies have shown that women generally outperform men in online learning environments.
A study by Dille (1991) revealed that students who completed a television broadcasting course were generally older, had higher GPAs, and possessed more college credits compared to dropouts Xenos (2002) found through field interviews that older students in separation instruction courses often required more encouragement from tutors, highlighting a critical relationship between age and dropout rates Additionally, Packham (2004) discovered that the majority of students who failed to complete their courses were male.
The online learning demographic is diverse, encompassing both young, tech-savvy students with part-time jobs and self-directed adults with clear objectives (Dabbagh, 2007) Research indicates that older students tend to achieve higher grades, aligning with findings that suggest they are generally more independent and self-motivated (Wojciechowski, 2005; Dabbagh, 2007; Knowles, 1989) These traits are crucial for success in an online learning environment Additionally, a study by Lim (2006) found that younger online learners, aged 20 to 29, not only performed better on knowledge assessments but also reported higher levels of satisfaction with their learning experience.
Research indicates that older students may struggle with epistemological skills and confidence in utilizing high-tech tools for web-based learning, as noted by Lim's findings and supported by Jones (1998) Additionally, age-related impacts on online learning are influenced by contextual factors and educational levels Shin (2004) found that students with higher education levels tend to exhibit greater confidence in their learning outcomes compared to those with lower educational backgrounds.
Students with prior experience in university education courses tend to have higher online completion rates There is a notable difference between those who successfully complete online courses and those who drop out, linked to their previous learning experiences Research indicates that students with higher education levels or longer periods of schooling are less likely to withdraw from online courses (Levy, 2007) However, some studies suggest that demographic factors have a minimal impact on dropout rates in distance education (Volkwein, 1995; Williamson, 1988).
H1a: Age has an indirect effect on intention to persist (P) through internal academic locus of control (ALOC)
H1b: Age has an indirect effect on intention to persist (P) through satisfaction (S)
Research indicates that older students tend to be more self-aware and focused on their goals, as noted by Dille (1991) These mature learners are less likely to engage in activities that do not contribute to their personal growth Furthermore, Fredericksen (2000) found that students who expressed the highest levels of satisfaction with various aspects of online courses also tended to be older, highlighting the connection between age, goal orientation, and online learning satisfaction.
Research indicates that older students often experience higher satisfaction levels in online courses compared to their younger counterparts This increased satisfaction can be attributed to their commitment to gaining knowledge for specific goals, leading to more focused study habits and better academic outcomes As a result, older learners are more likely to persist in their online courses Additionally, students with an internal locus of control tend to be more self-directed, motivated, and perform better than those with an external locus of control Older students, who generally possess a stronger internal locus of control, believe in their ability to achieve their goals, contributing to their persistence and success in online learning environments.
H1c: Gender has an indirect effect on intention to persist (P) through internal academic locus of control (ALOC)
H1d: Gender has an indirect effect on intention to persist (P) through satisfaction (S)
Research by Rovai (2001) indicates that men generally experience a lower sense of community compared to women, who exhibit a stronger feeling of connection within their networks This difference in connectivity often results in women experiencing less isolation and greater satisfaction, leading to increased persistence in online courses Conversely, men typically display higher confidence levels, particularly regarding chance or beliefs in higher powers, while women tend to have a greater inclination towards such beliefs Additionally, Adrian C Sherman (1997) found that men are more likely to possess an internal locus of control than women, influencing their perspectives on various situations.
35 influence of gender on intention to persist, men will have higher internal locus of control and get to be more perseverance in the online course
H1e: Educational level has an indirect effect on intention to persist (P) through internal academic locus of control (ALOC)
H1f: Educational level has an indirect effect on intention to persist (P) through satisfaction (S)
Higher educational levels correlate with enhanced knowledge and learning skills, leading to more effective learning and improved concentration Students with clear academic orientations experience greater satisfaction and demonstrate increased persistence in their studies As they accumulate academic experience and attain higher education, they better understand outcomes related to their abilities, fostering a strong internal locus of control Consequently, this leads to greater persistence in their educational pursuits.
Hypothesis 2: Internal factors have a direct, positive effect on intention to persist
H2a: Internal academic locus of control (ALOC) has a direct, positive effect on intention to persist (P)
The concept of academic locus of control refers to the belief that the results of one's actions are influenced either by personal efforts (internal control) or by external factors beyond one's influence (external control), as defined by psychologist Philip Zimbardo A study by Parker (1999) explored predictive factors for decision-making and perseverance in e-learning courses, emphasizing the role of demographic characteristics and locus of control in shaping student outcomes.
Parker's investigation identified locus of control as a key predictor of student dropout rates and academic diligence, achieving an accuracy rate of 80% Originating from Rotter's 1966 work, locus of control reflects an individual's awareness of the consequences of their actions in relation to how they perceive the outcomes of others' behaviors To assess this concept, Rotter developed a 40-item instrument designed to measure locus of control effectively.
Trice (1985) noted that while the Rotter (1966) instrument has been widely utilized, it has not effectively measured educational outcomes To address this gap, Trice developed a 28-item genuine bogus instrument based on Rotter's work to assess academic locus of control This concept allows students to reflect on their beliefs regarding their control over academic performance Additionally, Dollinger (2000) conducted a study examining the influence of academic locus of control on college course grades The findings revealed that students with a more internal locus of control achieved significantly higher grades in online courses compared to those with a more external locus of control.
Research indicates that students with an internal locus of control tend to be more self-directed, motivated, and perform better in online courses compared to those with an external locus of control (Chang, 2009; Liu, 2002; Parker, 2003) However, studies specifically examining academic locus of control are limited Lee (2012) found that students with an external academic locus of control have a higher dropout rate in e-learning courses In contrast, Levy (2007) reported that academic locus of control does not predict dropout rates in online education, highlighting inconsistencies in the findings Given the lack of research using updated scales, further investigation into the impact of locus of control on online student retention is essential.
Students with a high internal locus of control focus on their abilities and strive to study effectively rather than depending on luck They take responsibility for their actions and do not attribute their outcomes to external factors This mindset drives them to work diligently toward their goals, as they understand that their results are a direct reflection of their efforts and abilities, leading to greater persistence in their academic pursuits.
H2b: Satisfaction (S) has direct, positive effect on intention to persist (P)
RESEARCH METHODOLOGY
Research Process
Conclusion SEM CFA Factor analysis Conbach's Alpha analysis Collect survey officially Change Questionaire- Send Official Questionaire
Pilot Study - Translate Original Question into Vietnamese - Pre-test
Questionaire Literature Review Research Motivation
Data collection method
This thesis used primary data source (data that Master students collected)
Surveys were conducted to gather primary data on the key factors influencing the intention to persist in online courses in Vietnam This approach facilitated a comprehensive collection of relevant insights.
A survey conducted by Google Form targeted Vietnamese individuals aged 20 and above who were enrolled in various online courses, such as language, soft skills, and professional development, during the survey period from April 5 to April 20, 2020 The survey was exclusively distributed through online channels.
Due to the impracticality of surveying all online learners in Vietnam, this thesis centers on a sample of approximately 250 participants This sample size is determined using the calculation formula proposed by Hair (1998).
The research utilized Rovai's composite persistence model, necessitating the translation of the original research scale into Vietnamese A pre-test of the translated questionnaire was conducted with a sample group to ensure clarity and comprehension before administering the official survey.
This study utilized metrics adapted from previous research, including the Internal Academic Locus of Control metric modified from Trice (1985) and the Satisfaction metric derived from Keller's (1987) Instructional Materials Motivation Survey (IMMS) Additionally, support metrics were developed based on insights from the authors, Holder, focusing on family and work/school dynamics.
2007) Fiscal and Emotional Support and (Park, 2009), Persistence from Shin, N
(2003) After referring to the scale and questionnaire of previous studies, the questionnaire The survey for this study was formed with the following questions:
The study includes sixteen ladder questions ranging from disagreement to strong agreement, focusing on various aspects of online course learners in Vietnam It comprises three questions assessing internal academic locus of control, three questions evaluating learner satisfaction, six questions examining support from family and work/school, and four questions exploring the intention to persist in online learning.
+ Five (5) survey questions about demographic characteristics
Scale details are described in detail as follows:
Internal academic locus of control
ALOC 1: In my case, the good grades I receive are always the direct results of my efforts
ALOC 2: The most important ingredient in getting a good grade is my academic ability
ALOC 3: I feel that my good grades reflect directly on my academic ability
S1: It was a pleasure to work on this type of instruction
S2: I want to receive this type of instruction again for my studying improvement
S3: I enjoyed working with these user instructions so much that I was stimulated to keep on working
(Keller, 1987) (Instructional Materials Motivation Survey (IMMS)
Support from family and work/school(FW)
FW1: My family understands me very well even though I spend little time with them because of
FW2: My family is proud of me when I learn to improve my job/study performance
I will be able to meet my financial needs from my family while I pursue my study
FW4: My support group of colleagues/friends at my workplace/school encourages me to complete my program of study
FW5: My organization /school is willing to reduce my workload when I need training for my job/major
FW6: My supervisor/teacher shows interest in my job/major- related learning
Intention to persist (P) P1: Graduating from online course is important to me
P2: I am confident that I can overcome obstacles encountered in the course of online studying
P3: I will finish my studies at this online course no matter how difficult it may be
P4: I will certainly enroll for the next online course
- Design and evaluate the questionnaire:
The survey begins with a cover letter outlining the research's objectives and ethical considerations It consists of two sections: the first gathers demographic information and includes two questions regarding the respondents' most recent online course, categorizing them as completed, not completed, or currently learning Only responses from individuals actively participating in a course are considered valid for analysis The second section explores factors influencing the intention to persist in online courses, utilizing a five-point Likert scale to assess respondents' agreement with various statements, where 1 indicates "Strongly Disagree" and 5 indicates "Strongly Agree."
The questionnaire is designed to align with the objectives of the proposal and the hypothetical system in a clear and concise manner To ensure the accuracy of the survey, it will be organized using a specific methodology.
1 In light of the targets and research theoretical framework to recognize: aspects, variables and metrics
2 Decide the type of inquiry
3 Decide the substance of each question
4 Determine the wording utilized for each question
5 Decide the rationale for the inquiries
6 Trial survey and revision of questionnaire
8 Send the survey to the supervisor
9 Teachers check, modify and concur for send examination
A questionnaire was distributed to ten individuals participating in an online course to gather their feedback on the survey's question formulation and content While respondents had the opportunity to address all inquiries, some questions proved challenging to understand when translated into Vietnamese The author made revisions based on the evaluations provided by the participants.
RESEARCH FINDINGS
Descriptive Analysis
This part presents the investigation and related discoveries of all information gathered from the study Descriptive data analysis is a suitable technique to dissect elucidating survey review
A survey conducted using Google Forms reached over 400 individuals aged 20 and older in Vietnam Out of the total responses, 324 were collected, which included 27 dropouts and 63 completed surveys, leaving 234 qualifying answers for analysis.
- Age: 202 participants generally 86.3% are at the age of 20 to 29 years old, 27 participants generally 11.5% are at the age of 30 to 39 years old, otherwise 5 participants generally 2.2% are over 40 years old
The survey revealed the educational attainment of participants, showing that 3% held a high school diploma, 32% possessed a postgraduate degree, and 65% had completed a university degree.
- Gender: 69.2% of participants generally 162 people were female, 30.8% of participants generally 72 people were male.
Cronbach’s Alpha Analysis
The reliability testing of a measurement refers to the degree to which the instrument is free from random error, with reliability closely linked to the consistency and stability of the measurement This study utilized four independent scales—Demographic Characteristics, Internal Academic Locus of Control (ALOC), Satisfaction (S), and Support—alongside three dependent scales to assess the constructs of the Rovai revised model.
The study utilized three dependent scales: Intention to Persist (P), Internal Academic Locus of Control (ALOC), and Satisfaction (S), to effectively capture the significance of the proposed model To ensure the reliability and accuracy of the scales, a reliability test was conducted to evaluate both internal consistency and item-total correlations.
Internal consistency refers to the degree of reliability of responses within a single measurement scale (Kline, 2005) In this study, Cronbach's alpha was employed to assess internal consistency, as noted by Straub.
High correlations between elective proportions and substantial Cronbach's alphas indicate strong measure reliability (Straub, 1989) The evaluation of Cronbach's coefficient alpha is essential for assessing internal research consistency (Boudreau, 2004) According to the UTAUT model, a high internal consistency is expected, with a Cronbach's alpha value of 0.7 or above (Venkatesh, 2003) Hinton (2004) categorizes reliability into four levels: low (0.50 and below), moderate (0.50 to 0.70), high (0.70 to 0.90), and excellent (0.90 and above).
A reliability coefficient – Cronbach‟s alpha was run utilizing SPSS programming for set of develops
Table 4.1 Item statistic of intention to persist variable
Internal academic locus of control (ALOC)
Support from family and work/school (FW)
The Cronbach‟s Alpha proves all the variables: Internal academic locus of control, Satisfaction, Support from family and work/school and Intention to persist got Alpha
53 ratio with high reliability (>0.8) So that all of these variables are good enough to use analyze.
Factor Analysis
The investigation of disclosure factors prompts a focus on reducing the number of perceptions to a limited set of factors, allowing for a clearer analysis of the relationships among them An exploratory factor analysis (EFA) is deemed valid when it meets specific criteria.
+ Factor loading factor (Factor loading)> 0.5 According to Hair (1998, 111)
Differentiated load factors are defined as those exceeding 0.5 when considering a single factor In cases where two factors are analyzed, the absolute difference must be greater than 0.3 to be classified as significant Consequently, when filtering the matrix to display only load factors greater than 0.5, any invalid variables will be excluded To assess whether the standard difference in load factors between two groups surpasses 0.3, it is advisable to display coefficients of load factors that exceed this threshold within the software.
The KMO coefficient is a crucial index for assessing the adequacy of sample size in factor analysis, with values ranging from 0.5 to 1 indicating suitability A higher KMO value signifies a more appropriate sample size for conducting factor analysis, as noted by Mong (2007).
Bartlett's test indicates statistical significance (Sig < 0.050), suggesting that the factors in question are likely interconnected within the population When the test result is statistically significant (Sig < 0.05), it implies that the observed variables have a relationship, challenging the assumption that they are independent.
+ Eigenvalue values represent the the fluctuation clarified by each factor, just the factor with Eigenvalue more noteworthy than 1 will be held in the diagnostic model
+ Percentage of variance (Percentage of variance)> 50% This value shows the % change of variables which can be observed according to (Mong, 2007)
In this study, we conducted a Promax pivot analysis, revealing a Kaiser-Meyer-Olkin Measure of Sampling Adequacy of 0.885, indicating excellent sampling adequacy All observed factors exhibited loadings greater than 0.5, and the null hypothesis was rejected at a significance level of approximately 0% (Sig = 0.000), confirming the appropriateness of the exploratory factor analysis We identified 16 observed factors, which were grouped into 6 component clusters, with an Eigenvalue coefficient greater than 1 The total variance explained by these extracted factors was 64.339%, demonstrating that they account for a significant portion of the variability in the original data.
Table 4.3 KMO and Bartlett's Test
Kaiser-Meyer-Olkin Measure of Sampling Adequacy ,885 Bartlett's Test of Sphericity Approx Chi-Square 2282,554 df 120
Table 4.4 Eigenvalue and variance explained
Extraction Method: Principal Axis Factoring
Rotation Method: Promax with Kaiser Normalization a Rotation converged in 5 iterations
Confirmatory Factor Analysis
A factor investigation confirmed that confirmatory factor analysis (CFA) was conducted using 16 observed factors The results from the exploratory factor analysis (EFA) revealed four components, which were organized into scale groups that formed the applied measurement model These components were included in the CFA to evaluate the model's relevance to the research data The findings from the CFA analysis are as follows:
Figure 4-1 CFA model on Amos
4.4.1 Testing the suitability of the model
CMIN / DF = 2,191 ( 0.5 So that, measurement scales are relatively reliable
between family support and work/school satisfaction (FW and S) This indicates a strong relationship, with a correlation coefficient of 0.557 at the 5% significance level, supporting the validity of Hypothesis 4.
To establish convergent validity, a scale must have normalized loadings greater than 0.5 and be statistically significant (Gerbring and Anderson, 1988; Hair et al., 1992) Additionally, the Average Variance Extracted (AVE) should also exceed 0.5, as suggested by Fornell (1981) The study's findings indicate that all coefficients, both normalized and unnormalized, are above 0.5, and the AVE values also surpass this threshold Therefore, it can be concluded that the factors demonstrate convergent validity.
Table 4.7 Regression weights and standardized regression weights
According to Steenkamp (1991), the significance level of the model with the exploratory data provides essential conditions for achieving unidimensionality among the observed variables, independent of the measurement errors associated with them The results indicate that the model is consistent with the exploratory data, showing no correlation between measurement errors, which allows us to conclude that it is indeed unitary.
The discriminant value is evaluated by the below conditions:
Table 4.8 Assessment of Discriminant validity
The estimation of the square root of Average Variance Extracted (AVE) reveals that the relationship coefficients between different sets of elements do not equal 1 at a 95% confidence level.
The Fornell-Larcker criterion is a method used to assess the discriminant validity of latent variables by comparing the square root of the Average Variance Extracted (AVE) with the correlation coefficients of those variables For a factor to demonstrate adequate validity, the square root of its AVE must exceed the highest correlation coefficient it shares with any other factor This implies that the AVE should be greater than the square of the correlation coefficients with other factors, indicating that a factor captures more variance with its indicators than with other factors.
According to Table 4.5, the square root of the Average Variance Extracted (AVE) for each feature is greater than the correlation coefficients with other features, indicating that these ideas demonstrate strong validity.
Therefore, we have a suitable CFA investigation model
Structural Equation Modeling (SEM)
After CFA examination, the investigation utilized SEM structure model to decide the affecting elements and the degree of impact among the elements SEM
The investigation begins with an initial proposed research model, which is then refined to enhance its effectiveness Structural Equation Modeling (SEM) offers several advantages over traditional multivariate analysis techniques, such as multiple regression, by allowing for the calculation of measurement errors Additionally, SEM enables the integration of latent constructs with our measurements and accommodates both independent and dependent variables within the theoretical framework simultaneously.
Figure 4-2 Analysis results of SEM linear structure model
As being shown in figure 4.2, the model is logical with the research data because Chi square / df (2.342) 0.9;
RMSEA = 0.076 (